The perceived volume, or loudness, of different radio stations varies from one station to another. So-called mainstream stations such as Bayern 3 or Antenne Bayern [translator's note: local radio stations of the Federal State of Bavaria, Germany] are perceived to be considerably louder as compared to classical-music stations such as Bayern Klassik.
Nowadays, said differences are equalized by manual readjustment by means of the volume knob. In the long run, this may be annoying for the user. If radio is listened to inside a vehicle, this will even have relevance in terms of safety since the driver is inevitably distracted from the traffic situation while performing manual readjustment.
There have already been several approaches to (adaptively) adjusting the volume, or loudness, as a function of the signal. For example, the ITU standard BS.1770 describes a method of how aurally compensated calculation of loudness is to be effected. Disadvantages of said methods are that the audio signal is adapted during listening. Artefacts such as “pumping” (audible, fast volume adaptations), intense readjustments with plosives, or a reduction in the dynamics arise.
Due to their respective natures, said methods cannot meet the requirements placed upon transparent, unadulterated audio reproduction. In applications involving hifi requirements such as high-end audio systems in vehicles, for example, said algorithms therefore are not employed. Therefore, there is the need for an improved approach.